Publication number  US7002904 B2 
Publication type  Grant 
Application number  US 10/027,990 
Publication date  Feb 21, 2006 
Filing date  Dec 20, 2001 
Priority date  Jun 19, 2001 
Fee status  Lapsed 
Also published as  US20030067866 
Publication number  027990, 10027990, US 7002904 B2, US 7002904B2, USB27002904, US7002904 B2, US7002904B2 
Inventors  DaeKwon Jung 
Original Assignee  Samsung Electronics Co., Ltd. 
Export Citation  BiBTeX, EndNote, RefMan 
Patent Citations (4), NonPatent Citations (1), Referenced by (18), Classifications (9), Legal Events (4)  
External Links: USPTO, USPTO Assignment, Espacenet  
1. Field of the Invention
The present invention relates generally to reduction of a high peaktoaverage power ratio (PAPR) that is a problem with OFDM (Orthogonal Frequency Division Multiplexing), and in particular, to a partial transmit sequence (PTS) used for PAPR reduction.
2. Description of the Related Art
OFDM is used as a transmission technique for digital broadcasting and will be a significant technology for future high rate multimedia wireless communications. A serious problem with OFDM is high peak power, and present known methods are deficient in providing adequate techniques that result in the reduction of the high peak power.
OFDM is a type of multicarrier modulation scheme that shows excellent performance in a multipath fading mobile communication environment. For this reason, OFDM attracts interest as a modulation scheme suitable for ground wave digital video broadcasting (DVB) and digital audio broadcasting (DAB). In this modulation scheme, low rate transmission is performed in parallel on multiple carriers instead of high rate transmission on a single carrier. With parallel transmission, the symbol period of multiple carrier signals is longer than that of original input data. Thus, fading does not focus on a few adjacent bits causing errors but slightly influences a plurality of parallel bits. The robustness to channel fading makes OFDM less sensitive to channel distortion.
OFDM shows a high spectral efficiency because the spectrums of subchannels are orthogonal and overlap each other. Since OFDM modulation/demodulation is implemented by IDFT/DFT (Inverse Discrete Fourier Transform/ Discrete Fourier Transform), efficient digitalization of a modulator/demodulator results. Despite these advantages, OFDM has an inherent shortcoming of a high PAPR in multicarrier modulation. A signal with a high PAPR decreases the efficiency of a linear amplifier and sets an operational point in a nonlinear area for a nonlinear amplifier, resulting in mutual modulation between carriers and spectrum distribution. Many methods have been suggested to reduce the PAPR.
The PAPR is defined as
where s(nT) is a signal sampled in intervals T, x_{n }is the sample value of the signal, and E{·} is an operator for calculating an expected value.
A basic PAPR reduction method is to operate a linear amplifier in a linear area or to operate a nonlinear amplifier in a linear area by use of a backoff. However, this method has the disadvantage of low amplifier efficiency. More simply, clipping can be used whereby if the strength of a signal is greater than a predetermined value, the difference is clipped. Clipping, furthermore, deteriorates BER (Bit Error Rate) performance because of the band distortion caused by nonlinear operation. In addition, clipping generates noise in the band, decreasing spectral efficiency.
Coding also can be applied to multiple carriers for the PAPR reduction method. Extra carriers are transmitted with parity bits to reduce the PAPR. This method remarkably reduces the PAPR due to its error correction function and an additional narrow bandwidth. Yet, as more carriers are used, the size of a lookup table or a generation matrix increases, thereby increasing hardware complexity, reducing operation efficiency, and decreasing operational speed.
Two schemes showing promising PAPR reduction have recently been proposed.
One is selective mapping (SLM) and the other is partial transmit sequence (PTS). According to the SLM scheme, data of length N is multiplied by M statistically independent sequences of length N and a sequence with the lowest PAPR is transmitted selectively. Despite the requirement of M IFFT (Inverse Fast Fourier Transform) processes, the SLM scheme can reduce a PAPR greatly and be applied irrespective of the number of carriers.
On the other hand, the PTS scheme, having the advantages of the SLM scheme, is known as the most effective flexible method of reducing the PAPR without nonlinear distortion.
The PTS scheme applies flexibly according to its subblock partition methods: interleaved subblock partition, adjacent subblock partition, and pseudorandom subblock partition. According to the interleaved subblock partition method shown in
where N is the number of carriers where the partitioned data block is assigned (L=N/M) and M is the number of subblocks.
In the adjacent subblock partitioning method shown in
In the pseudorandom subblock partitioning method shown in
Table 1 below lists computation volumes for the above three subblock partitioning methods when N=256 and M varies.
TABLE 1  
A Comparison between Complex Computation Volumes for Three  
Subblock Partitioning Methods of PTS  
Interleaved  Adjacent, Pseudorandom  
M  Multiplication  Addition  Multiplication  Addition 
1  1280  2048  1024  2048 
2  1408  1792  2048  4096 
4  1792  1536  4096  8192 
8  2688  1280  8192  16384 
16  4608  1024  16384  32768 
Concerning computation volume and performance, the concatenated pseudorandom subblock partitioning method is between the interleaved and pseudorandom methods though its computation volume and performance differ depending on the concatenation factor C. The computation volume and performance are in a tradeoff relation. Therefore, this concatenated pseudorandom subblock partitioning method increases the scope of choice for the PTS scheme according to required performance and computation volume.
Since each partitioned data block of length L divided from the input data block of length N is assigned to the corresponding M subblocks with “0” values, it increases the number of the optimizing operation as much as M*N and the amount of computation for IFFT. Accordingly, it causes an increase of the high peaktoaverage power ratio (PAPR) for the conventional art.
It is, therefore, an object of the present invention to provide a method and apparatus of reducing computation volume and increasing PAPR reduction performance by decreasing the number of PAPR optimizing operations.
It is another object of the present invention to provide a method and apparatus of reducing interference between data by multiplying data output by an orthogonal coefficient.
The above and other objects can be achieved by providing a novel subblock partitioning method for PAPR reduction in a PTS OFDM scheme. In an OFDM transmitting apparatus for peak power reduction according to the present invention, an interleaved subblock partitioner partitions an input data block of length N into M, the number of subblock, and distributes the data of each partitioned data block in M subblocks one by one rotationally. M IFFTs perform N/Mpoint IFFT on N/M data assigned to each of the M subblocks and M coefficient multipliers multiply the N/M data output from each IFFT by corresponding a predetermined coefficients to give orthogonality to the frequency components of the N/M output data. A phase factor optimizer optimizes M phase factors to minimize a PAPR using the N/M output values of each coefficient multiplier. M multipliers multiply the optimized M phase factors by the outputs of the coefficient multipliers. A transmitter sums the outputs of the M multipliers on a symboltosymbol basis in the same positions and transmits the sum.
The above and other objects, features and advantages of the present invention will become more apparent from the following detailed description when taken in conjunction with the accompanying drawings in which:
A preferred embodiment of the present invention will be described hereinbelow with reference to the accompanying drawings. In the following description, wellknown functions or constructions are not described in detail since they would obscure the invention in unnecessary detail.
While interleaved subblock partitioning is optimum for the PTS scheme in computation volume, pseudorandom subblock partitioning shows the best performance in PAPR reduction. Thus, the PAPR reduction performance and the computation volume are in a tradeoff relation. For clarity of description, a subblock partitioning method for the PTS scheme according to the present invention will be termed “InterleaveandTransmitLsample” (hereinafter, ITL).
A mathematical analysis of the ITL method is set forth below. Considering the ITL method, the elements of each vector X^{(m) }(0≦m≦M−1) on the basis of the conventional interleaved subblock partitioning have periodicity expressed as
where L=N/M. By applying an IFFT formula to each subblock, a time domain signal is given by
As seen from Eq. (5), for n=0, L, 2L, . . . ,
is the same. The case is also true for n=1, L+1, 2L+1, . . . In the interleaved subblock partitioning method, therefore, Npoint IFFT is equivalent to performing Lpoint IFFT on L nonzero data of the vector X^{(m)}, repeating the Lpoint IFFT values M times, multiplying the repeated values by coefficients
and summing the multiplication results. In other words, the Npoint IFFT can be expressed as repeating L time domain samples M times and multiplying the repeated Lpoint IFFT samples by
before final time domain samples are achieved. Table 2 lists coefficients to be multiplied in the receiver, on the other hand the inverse number of the coefficients listed Table 2 to be multiplied in the transmitter to give orthogonality to the frequency components of the output data.
TABLE 2  
Coefficients for Multiplication in the Receiver  
m\n  0  1  2  . . .  L − 1 
0  1  1  1  . . .  1 
1  1 


. . . 

2  1 


. . . 

. . .  . . .  . . .  . . .  . . .  . . . 
M − 1  1 


. . . 

To implement the scheme presented by Eq. (5), M IFFTs (Inverse Fast Fourier Transformers) perform Lpoint IFFT on the N/M data in each subblock. M coefficient multipliers multiply N/M data received from each IFFT by the corresponding coefficients to give orthogonality to the frequency components of the output data of each M IFFT since the output data being overlaid between each subblock have the same frequency components. The coefficient can be
and its absolute value can be changed. The optimizer optimizes M phase factors using a predetermined optimization algorithm. The optimized M phase factors are multiplied by the outputs of the coefficient multipliers. Then, the products are summed on a symbol basis by summing sample values at the same time point, thereby achieving a minimized PAPR.
A phase factor eliminator 710 multiplies the data of each subblock of length L assigned in the rotational manner. Phase factors are optimized and transformed, using the outputs of each coefficient multiplier according to a predetermined PAPR optimal algorithm by the optimizer 640. The phase factors are multiplied by the outputs of each coefficient multiplier 630 by the corresponding multipliers 650. The outputs of each multiplier 650 multiplied by the output of the corresponding phase factors and the coefficient multipliers 630 are summed bit by bit in the adder 660. Accordingly, the increased PAPR reduction can be achieved by adding each output of the multipliers 650 due to the decrease number of samples from N to L and the lesser amount of the PAPR optimizing operations.
To implement the scheme presented by Eq. (5), M IFFTs (Inverse Fast Fourier Transformers) perform Lpoint IFFT on the N/M data in each subblock. M coefficient multipliers multiply N/M data received from each IFFT by the corresponding coefficients to give orthogonality to the frequency components of the output data of each M IFFT since the output data being overlaid between each subblock have the same frequency components. The coefficient can be
and its absolute value can be changed. The optimizer optimizes M phase factors using a predetermined optimization algorithm. The optimized M phase factors are multiplied by the outputs of the coefficient multipliers. Then, the products are summed on a symbol basis by summing sample values at the same time point, thereby achieving a minimized PAPR.
A phase factor eliminator 710 multiplies the data of each subblock of length L by a corresponding inverse phase factor. M inverse coefficient multipliers 720 use e^{−j2πmn/N }(0≦m≦M−1, 0≦n≦L−1) to remove orthogonality from the frequency components of L signals in each subblock. M FFTs (Fast Fourier Transformers) 730 perform Lpoint FFT on the signals received from the inverse coefficient multipliers. A deinterleaver 740 deinterleaves the signals received from the Lpoint FFTs, thereby recovering the original signal.
The transmitter obtains L consecutive time domain samples for each subblock by one Lpoint IFFT operation and L complex multiplications. Since demodulation is carried out in the reverse order of the modulation, the receiver demodulates each subblock by L complex multiplications and one Lpoint FFT. Therefore, the requirement of computation for the transmitter and receiver is calculated by
Let us compare the conventional interleaved subblock partitioning method with the present invention in terms of computation volume, keeping in mind that CooleyTuckey IFFT is applied to modulation related with the interleaved method and typical Npoint FFT is used for demodulation. Table 3 shows the comparison results. Here, the number of subcarriers per symbol is 256 and the number of subblocks varies up to 16.
TABLE 3  
A Comparison between the Interleaved Method and the ITL Method  
in terms of Complex Computation  
Interleaved  ITL  
Transmitter  Receiver  Transmitter  Receiver  
Multipli  Multipli  Multipli  Multipli  
M  cation  Add  cation  Add  cation  Add  cation  Add 
1  1280  2048  1024  2048  1280  2048  1280  2048 
2  1408  1792  2048  4096  1152  1792  1152  1792 
4  1792  1536  4096  8192  1024  1536  1024  1536 
8  2688  1280  8192  16384  896  1280  896  1280 
16  4608  1024  16384  32768  768  1024  768  1024 
As noted from Table 3, as M increases, computation volume decreases in the ITL method and increases in the interleaved method. For M=8, complex multiplications for the transmitter in the ITL method are about an eighth of those in the interleaved method and complex multiplications for the receiver in the ITL method are about a ninth of those in the interleaved method. For M=16, the difference in computation volume between the interleaved method and the ITL method gets wider. Hence, it can be said that the ITL method is superior to the interleaved method as well as the other conventional methods in computation volume. Now, a description will be made of improved system performance in the present invention.
It is noted from
Therefore, optimized phase factors differ in the ITL method and the conventional methods. This leads to a big difference in PAPR between the methods. A survey of time domain samples in the ITL method will show how much the ITL method improves system performance.
While recent research efforts have been expended toward the pseudorandom subblock partitioning method having a high computation requirement but exhibiting the best performance among the three subblock partitioning methods for the PTS scheme, the ITL subblock partitioning method shows far better PAPR reduction performance than the interleaved subblock partitioning method and the pseudorandom subblock partitioning method. Furthermore, since a transmitter/receiver according to the present invention requires a less computation volume than that of the conventional methods, hardware complexity is reduced and power consumption become low, which makes the transmitter/receiver for the present invention suitable for a mobile terminal. Considering that OFDM will be adopted increasingly for wide usage in wireless communication applications as well as wired communication applications, the present invention is very significant in that it is effective in PAPR reduction being a conventional challenging issue in many studies.
While the invention has been shown and described with reference to a certain preferred embodiment thereof, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.
Cited Patent  Filing date  Publication date  Applicant  Title 

US6314146 *  Jun 5, 1998  Nov 6, 2001  The Board Of Trustees Of The Leland Stanford Junior University  Peak to average power ratio reduction 
US6424681 *  Apr 20, 1998  Jul 23, 2002  The Board Of Trustees Of The Leland Stanford Junior University  Peak to average power ratio reduction 
US6512797 *  May 19, 1998  Jan 28, 2003  The Board Of Trustees Of The Leland Stanford Junior University  Peak to average power ratio reduction 
US6556557 *  Jun 2, 1999  Apr 29, 2003  At&T Corp.  Method and system for reducing of peaktoaverage power ratio of transmission signals comprising overlapping waveforms 
Reference  

1  *  Muller S. H.; Huber J.B.; "OFSM with reduced peaktoaverage power ratio by optimum combination of partial transmit sequences", Feb. 27, 1997, Electronics Letters, vol. 33, pp. 368369. 
Citing Patent  Filing date  Publication date  Applicant  Title 

US7227903 *  Jul 22, 2003  Jun 5, 2007  Mitsubishi Electric Research Laboratories, Inc.  OFDM transmitter for generating FSK modulated signals 
US7733976 *  Aug 31, 2006  Jun 8, 2010  James Stuart Wight  Computational circuits and methods for processing modulated signals having nonconstant envelopes 
US7990840  Aug 2, 2011  Qualcomm Incorporated  Methods and apparatus for efficient digital broadcast signaling in a wireless communications system  
US8068558  Nov 29, 2011  Nortel Networks Limited  Selective peak power reduction  
US8175177 *  Aug 13, 2008  May 8, 2012  Lg Electronics Inc.  Peak to average power ratio reduction 
US8238473  Aug 7, 2012  Research In Motion Limited  Selective peak power reduction  
US8582687  Jun 28, 2010  Nov 12, 2013  Plusn, Llc  System and method for controlling combined radio signals 
US8787873  Nov 5, 2012  Jul 22, 2014  Plusn Llc  System and method for communicating using bandwidth on demand 
US9160593  Nov 5, 2013  Oct 13, 2015  Plusn Llc  System and method for controlling combined radio signals 
US20050018783 *  Jul 22, 2003  Jan 27, 2005  YvesPaul Nakache  OFDM transmitter for generating FSK modulated signals 
US20060223520 *  Mar 7, 2006  Oct 5, 2006  Rajiv Laroia  Methods and apparatus for efficient digital broadcast signaling in a wireless communications system 
US20070073796 *  Sep 18, 2006  Mar 29, 2007  Newlogic Technologies Ag  Method and apparatus for fft computation 
US20070201353 *  Aug 31, 2006  Aug 30, 2007  Wight James S  Computational circuits and methods for processing modulated signals having nonconstant envelopes 
US20090052577 *  Aug 13, 2008  Feb 26, 2009  Lg Electronics Inc.  Peak to average power ratio reduction 
US20100150256 *  Dec 17, 2008  Jun 17, 2010  Nortel Networks Limited  Selective peak power reduction 
US20100329401 *  Jun 28, 2010  Dec 30, 2010  Hypres, Inc.  System and method for controlling combined radio signals 
WO2010070406A2 *  Dec 2, 2009  Jun 24, 2010  Nortel Networks Limited  Selective peak power reduction 
WO2010070406A3 *  Dec 2, 2009  Oct 28, 2010  Nortel Networks Limited  Selective peak power reduction 
U.S. Classification  370/210, 370/208, 370/480, 375/260 
International Classification  H04J11/00, H04M3/42, H04L27/26 
Cooperative Classification  H04L27/2614 
European Classification  H04L27/26M2 
Date  Code  Event  Description 

Apr 16, 2002  AS  Assignment  Owner name: SAMSUNG ELECTRONCIS CO., LTD., KOREA, REPUBLIC OF Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:JUNG, DAEKWON;REEL/FRAME:012823/0959 Effective date: 20020403 
Sep 28, 2009  REMI  Maintenance fee reminder mailed  
Feb 21, 2010  LAPS  Lapse for failure to pay maintenance fees  
Apr 13, 2010  FP  Expired due to failure to pay maintenance fee  Effective date: 20100221 